Desktop User Guides > Author > UNICOM Intelligence Question Repository > UNICOM Intelligence Question Repository architecture > UNICOM Intelligence Question Repository organization > Search fields in a UNICOM Intelligence Question Repository
 
Search fields in a UNICOM Intelligence Question Repository
When a survey asset .mzd file (see Survey assets in a UNICOM Intelligence Question Repository) is stored to the repository, IBM SPSS Collaboration and Deployment Services indexes the content of all .mdd files inside the .mzd file (including the supplemental .mdd file that is used to store attributes (see Attributes in a UNICOM Intelligence Question Repository) but ignores any other file types. All created index entries refer to the .mzd file.
When IBM SPSS Collaboration and Deployment Services moves a file between folders, the existing index entries for that file are deleted and the file is indexed again so that the index entries reflect the new location.
Do not use IBM SPSS Deployment Manager to move, copy, update, delete, or rename existing .mzd files, or to delete, rename, or import folders that contain .mzd files. This method does not update the indexes used to search for these files. However, you can use IBM SPSS Deployment Manager to add new files and folders.
What is indexed
If the metadata document in the .mzd file is versioned, only the latest version is indexed. Index entries are created for the following metadata items:
the names and label texts for all questions, categories, and shared lists. When label texts have been specified for more than one combination of context and language, an index entry is created for each alternative label text.
the names of all contexts, languages, and routing contexts
the names and values of the document's custom properties.
In addition, index entries are created to record the following:
the name of the repository folder in which the asset is currently stored (see Folders in a UNICOM Intelligence Question Repository)
whether the file contains any questions
whether the file contains any shared lists.
The value for the last two index entries is set to either True or False.
What is not indexed
The following items are not indexed:
system variable names and label texts
the names and label texts of loop, grid, compound, block, and page questions. However, simple questions inside loops, grids, compounds, and blocks are indexed, as are the names and label texts of the controlling categories in a categorical loop or grid.
label texts that are inherited from context or language alternatives
custom properties for any object other than the document object.
See also
UNICOM Intelligence Question Repository organization